The Genomic Organization of Plant Pathogenicity in Fusarium Species Martijn Rep1 and H Corby Kistler2

Available online at www.sciencedirect.com

The genomic organization of plant pathogenicity in Fusarium species Martijn Rep1 and H Corby Kistler2

Comparative genomics is a powerful tool to infer the molecular cation and genetic transmission. We here highlight recent basis of fungal pathogenicity and its evolution by identifying insights into the evolution of disease-causing ability differences in gene content and genomic organization between among plant pathogenic fungi, focusing on the compara- fungi with different hosts or modes of infection. Through tive genomic analysis of Fusarium species with additional comparative analysis, pathogenicity-related chromosomes reference to other fungi. have been identified in Fusarium oxysporum and Fusarium solani that contain genes for host-specific virulence. Lateral Comparative genomics transfer of pathogenicity chromosomes, inferred from genomic In 2007 the Broad Institute released its first Fusarium data, now has been experimentally confirmed. Likewise, comparativegenomicswebsite(http://www.broadinstitute. comparative genomics reveals the evolutionary relationships org/annotation/genome/fusarium_group/MultiHome. among toxin gene clusters whereby the loss and gain of genes html), which brought together high quality sequence from the cluster may be understood in an evolutionary context assemblies of the plant pathogenic fungus Fusarium grami- of toxin diversification. The genomic milieu of effector genes, nearum, sequenced previously [1] and two new WGS for encoding small secreted proteins, also suggests mechanisms the species Fusarium verticillioides and Fusarium oxysporum. that promote genetic diversification for the benefit of the At the sametime,theJoint Genome Institute (JGI) released pathogen. a WGS for Fusarium solani (Nectria haematococca)(http:// Addresses genome.jgi-psf.org/Necha2/Necha2.home.html). The four 1 Plant Pathology, Swammerdam Institute for Life Sciences, Faculty of genomes share considerable sequence similarity as well as Science, University of Amsterdam, P.O. Box 94215, 1090 GE extensive synteny (Figure 1)[2,3]. Amsterdam, The Netherlands 2 USDA ARS Cereal Disease Laboratory, 1551 Lindig St., University of Minnesota, St. Paul, MN 55108, USA The Fusarium genomes consist of a core region with approximately 9000 genes considered to be orthologous Corresponding authors: Rep, Martijn ([email protected]) and Kistler, H Corby due to high sequence similarity and conserved gene order ([email protected]) [3]. Each species also contains thousands of genes that are unique to each genome, many of which are found near Current Opinion in Plant Biology 2010, 13:420–426 the ends of chromosomes. In F. graminearum, these telomere proximal regions are rich in gene diversity as This review comes from a themed issue on measured by SNP density [1] and are regions of elevated Biotic interactions Edited by Jane E. Parker and Jeffrey G. Ellis recombination [4]. Distinctively, F. graminearum also contains interstitial chromosomal regions of high diversity Available online 13th May 2010 and recombination that appear to have been created by an 1369-5266/$ – see front matter ancestral telomeric fusion of chromosomes [1 ,3 ]. How Published by Elsevier Ltd. these interstitial regions of F. graminearum chromosomes have maintained high genetic diversity, normally associ- DOI 10.1016/j.pbi.2010.04.004 ated with telomeres, remains a mystery.

Comparative genomics with closely related Aspergillus Introduction species also has revealed large species-specific gene sets, Genomic sequencing of fungal phytopathogens has revo- which are concentrated in subgenomic regions called lutionized the study of plant pathogenesis. Whole gen- chromosomal islands [5 ]. Interestingly, these regions ome sequence (WGS) data for individual fungal genomes have a subtelomeric bias too. This could be due to the accelerated classical forward and reverse genetic inherent recombinogenic nature of chromosome ends but approaches for identifying pathogenicity genes. More seems to have functional implications as well; subtelo- recently, the availability of several WGS assemblies for meric gene expression is associated with invasive growth comparative genomic analysis has enabled unprece- of Aspergillus fumigatus in mammals and ex vivo neutrophil dented opportunities for tracing the evolutionary origin exposure [6]. (and demise) of genes and molecules that influence the outcome of fungal–plant interactions. Moreover, the over- Instead of genomic islands, both F. oxysporum and F. all genomic organization of fungal pathogenicity-related solani contain supernumerary chromosomes that largely genes has suggested novel modes of molecular diversifi- consist of lineage-specific sequences and make up a

Current Opinion in Plant Biology 2010, 13:420–426 www.sciencedirect.com Genomic organization of plant pathogenicity in Fusarium species Rep and Kistler 421

Figure 1

Comparison of Fusarium Genomes. Synteny of the F. oxysporum, F. verticillioides, F. graminearum,andF. solani genomes determined by BLASTN alignment (cutoff 1e-10). Chromosomes (Chr) of F. oxysporum are used as a reference and the scale (bottom) measures chromosome size in megabases (Mb). Grey boxes indicate chromosome size determined by optical mapping, upon which are superimposed corresponding WGS assembly scaffolds displayed as numbered black bars. Unassembled scaffolds (Un) are concatenated for ease of display. Regions of shared gene order with chromosomes of F. verticillioides, F. graminearum and F. solani, are shown beneath each F. oxysporum chromosome. Sequences from individual chromosomes in each species are color-coded so that blocks of solid color represent regions of synteny with the F. oxysporum genome. Little homology or synteny is apparent for the lineage specific regions of F. oxysporum chromosomes 3, 6, 14 and 15 and for telomere-proximal portions of chromosomes 1 and 2. Figure modified from supplemental material of [3]. www.sciencedirect.com Current Opinion in Plant Biology 2010, 13:420–426 422 Biotic interactions

substantial fraction of the entire genome [2,3]. Lin- co-cultivation, resulting in a new pathogenic lineage. eage-specific portions of F. oxysporum and F. solani are Lateral chromosome transfer, first observed in Colleto- highly enriched in transposable elements. Regions of high trichum gloeosporioides [17], likely explains the existence genome diversity in Fusarium, whether at chromosome of host-specific groups of strains ( formae speciales)ofF. ends, special interstitial sites or on supernumerary oxysporum composed of several independent lineages. chromosomes, also have been shown to be rich in genes This is supported by the observation that the sequences predicted to be involved in fungal–host interactions. High of the effector genes on the pathogenicity chromosome diversity regions are enriched for predicted secreted are, with few exceptions, identical between disparate proteins, carbohydrate-active enzymes and genes specifi- lineages and highly specific to lineages which are patho- cally expressed during plant infection [1,3,7]. genic to tomato [18].

Effector genes, pathogenicity chromosomes In the pea-pathogenic form of F. solani, supernumerary and horizontal transfer chromosomes also harbor genes for host-specific viru- Effectors are proteins secreted by pathogens that promote lence, including detoxification mechanisms and host virulence, commonly by interacting with plant host nutrient utilization [19,20]. Three supernumerary proteins [8]. Because of these interactions, effector genes chromosomes with sizes between 0.5 and 1.6 Mb have are frequently involved in molecular arms races between characteristics that distinguish them from the rest of the pathogen and plant and subject to accelerated evolution genome including higher levels of repeat sequences [9]. The location of effector genes in a genome may affect (mostly transposons), more duplicated and unique genes, the rate at which they evolve through mutation or recom- smaller average gene size and a lower G+C content. Like bination. In Leptosphaeria maculans, for instance, the in F. oxysporum, sequences of the genes on these chromo- effector (Avr) genes that have been identified through somes suggest activities consistent with habitat special- positional cloning all reside in AT-rich genomic subre- ization but no involvement in essential cellular functions gions of up to several hundred kb consisting mostly of [2]. For Alternaria alternata, too, host plant-specific remnants of retrotransposons [10,11,12]. These long pathogenicity can be attributed to supernumerary chromosomal segments of uniform AT/GC content are chromosomes, of 1.0–1.7 Mb [21]. Genes for synthesis similar to isochores found in other eukaryotes. of host-selective toxins that are crucial for host-specificity are located on these chromosomes [22]. Although these highly particular, isochore-like genomic subregions presently have been found only in L. maculans, Transposons the genomic location of effector genes is also non-random Could the proximity of effector genes to repeats or in several other fungi. In the corn pathogen Ustilago transposons accelerate their evolution? While Lepto- maydis, a significant fraction (18%) of the genes for sphaeria represents an extreme case of repetitive genomic secreted proteins are organized in 12 clusters (3–26 genes context of effector genes, the context of effector genes in per cluster) dispersed in the genome [13]. Most clusters other fungi is also often transposon rich [16,23]. In contain tandem arrays of 2–5 related genes. This is Leptosphaeria, both transposons and effector genes appear remarkable since overall, the genome of U. maydis is to undergo mutation through Repeat Induced Point largely devoid of repetitive DNA. Expression of most Mutation (RIP) in the AT-rich genomic subregions genes in the U. maydis effector gene clusters is co- [24]. In F. solani, 72% of the repetitive sequences but regulated and induced during plant infection [13]. only 4% of the unique sequences have been subjected to RIP [2]. It will be interesting to determine whether a In the rice blast fungus, Magnaporthe oryzae, effector genes repetitive genomic environment promotes mutation of are not clustered but are frequently found near telomeres, non-repetitive effector genes residing within it. Acceler- where gene duplication and gene conversion may be more ated DNA sequence divergence coupled with positive frequent [14,15,16]. One example of an apparent recent selection for effector diversification might link transpo- telomeric duplication in Fusarium is the effector gene sons to the evolution of nearby effector genes. A repeat- SIX8 in F. oxysporum, which is present in identical copies rich genomic environment might also promote homolo- close to at least eight telomeres in the genome of the gous recombination among different repeats leading to tomato-specific pathogenic strain (M Rep, LJ Ma, unpub- duplication, chimeric gene formation or gene loss. In line lished observations). with this, transposons are associated with gene gain/loss events in M. oryzae [16]. Also, transposons are known to Interestingly, all other known effector genes in the mediate chromosomal rearrangements [25] and to affect tomato-pathogenic strain of F. oxysporum are localized chromatin structure [26], which may influence effector on a single, transposon-rich ‘pathogenicity chromosome’ gene expression and possibly also mutation rates. of 2Mb[3]. This chromosome and another smaller strain-specific chromosome can undergo transfer be- The pathogenicity-related and other supernumerary tween pathogenic and non-pathogenic strains during chromosomes that have been analyzed are all enriched

Current Opinion in Plant Biology 2010, 13:420–426 www.sciencedirect.com Genomic organization of plant pathogenicity in Fusarium species Rep and Kistler 423

for transposable elements. Interestingly, in a clear case of Recent comparative genomics of the trichothecene gene horizontal gene transfer between fungi, a single gene cluster among Fusarium species has captured the process encoding a host-selective toxin was transferred together of how genetic elements are added and subtracted from with an hAT-family DNA transposon [27]. It remains to clusters resulting in different chemotypes (Figure 2). In be determined whether association of transposons with F. graminearum the trichothecene gene cluster consists of mobile virulence-related genes or whole chromosomes is 10–12 contiguous genes as well as two other genes, Tri1 coincidental, resulting from ‘tolerance’ of non-essential and Tri101, which are at separate loci outside the main genomic regions to transposon insertion, or whether cluster [33]. While the intact gene cluster in F. grami- transposons are selected for function in the transfer nearum results in strains that produce the toxic compound process. For now this remains mostly speculative; mech- NIV, naturally occurring alternative forms of the gene anisms through which transposons may promote horizon- cluster exist for different chemotypes [29,34] that have tal transfer of genes or chromosomes are currently pseudogenes or deletions for Tri7 and Tri13. Tri13 is a unknown. cytochrome P-450 monooxygenase responsible for the 4- position hydroxyl that distinguishes NIV and DON che- Evolution of toxin diversity at biosynthetic motypes. Targeted mutation of the Tri13 gene in the NIV gene clusters cluster therefore results in strains that produce DON [35]. Fungal genes for biosynthesis of secondary metabolites, including toxic compounds produced in plants, often are Further diversification occurs for trichothecene chemo- clustered at a single locus and are co-expressed [28]. Two types in species distant to F. graminearum. F. sporotri- major themes in toxin cluster evolution have been chioides and related species produce A-type revealed by comparative genomics: trans-species poly- trichothecenes, such as T-2 toxin, which differs from morphism and linkage disequilibrium within blocks of NIV and DON by having an isovalerate ester at the 8 genes, each correlated to ‘chemotype’ [29–31]. Chemo- position oxygen, rather than a carbonyl. These differ- types are distinct spectra of metabolites/toxins produced ences result from catalytic divergence of the cytochrome by related strains or species. Examples of toxin chemo- P-450 enzymes encoded by Tri1.InF. graminearum Tri1p types are the nivalenol (NIV)- versus deoxynivalenol oxygenates both C-7 and C-8 (which results in a hydroxyl (DON)-producing strains of Fusarium. Polymorphisms at C-7 and a carbonyl at C-8) whereas in F. sporotrichioides, in multiple genes in the toxin pathway lead to separate only C-8 is hydroxylated by Tri1p [36,37]. In F. grami- chemotypes. As selection is likely occurring on the toxic nearum, Tri1 is separate from the main trichothecene gene pathway end-product(s), recombination between genes cluster at a telomere-proximal, high diversity region of the that define different chemotypes might result in combi- genome, whereas the main cluster is at a low diversity nations that produce no toxins or a level or spectrum of genomic position. We speculate that the diversification toxin with reduced biological activity. Negative selection for chemotype made possible by the divergence of Tri1 against recombination of genes defining different che- was facilitated by its position near the telomere. Like the motypes might, therefore, underlie the observed pat- cluster itself, Tri1 also exhibits trans-species polymorph- terns of linkage disequilibrium across the gene cluster ism that appears to track with chemotype [33]. However, [29,30]. in Fusarium equiseti and related species, Tri1 has relocated to the main trichothecene cluster along with Tri101 and a Trans-species polymorphisms correlated with chemotype gene for a predicted transcription factor [33]. The vari- suggest that cluster diversification arose prior to species ation generated at a high diversity genomic locus may be divergence. This interpretation is supported by coalesc- ‘locked in’ by migration to the cluster where diversifica- ence analysis of toxin cluster and non-cluster genes in tion and recombination may be limited. Aspergillus whereby cluster genes display a twofold greater difference in the estimated time to the most common The sporadic phylogenetic distribution of secondary recent ancestor [31]. Trans-species polymorphism may be metabolite gene clusters among fungal species initially due to balancing selection, and so it is tempting to led to the hypothesis that many gene clusters could have speculate how balancing selection may explain the per- been horizontally transferred among species as they have sistence of toxin chemotypes. One possibility is that been in bacteria (reviewed in [38]). The phylogeny of chemotypes may confer differential fitness in different genes within clusters, however, often is congruent with hosts. Trichothecenes produced by F. graminearum are the phylogeny inferred from housekeeping genes. Most known to alter the pathogenicity of the fungus in complex cluster evolution therefore seems to occur by way of and host-specific ways. For example, the trichothecene vertical transmission, with cluster duplication and diver- NIV, but not DON, is required for maximum aggressive- sification combined with frequent cluster loss resulting in ness on maize but not on wheat [32]. If differences in a pattern resembling horizontal transfer [39]. Dis- disease level reflect fitness differences, then chemotype tinguishing between vertical and horizontal transmission diversity may be maintained due to differential fitness of has been facilitated by DNA sequence information from chemotypes in the various hosts of Fusarium. a wide range of filamentous fungi, which has enabled www.sciencedirect.com Current Opinion in Plant Biology 2010, 13:420–426 424 Biotic interactions

Figure 2

Comparison of trichothecene biosynthetic gene clusters from different Fusarium species and chemotypes. Clusters are shown for F. equiseti,and (top to bottom) F. graminearum NIV, 15 ADON and 3 ADON chemotypes. Arrows in orange represent trichothecene biosynthetic (Tri) genes with gene numbers given below. Arrows in green are Tri genes that have moved to the Tri cluster in F. equiseti. T is a predicted transcription factor. Arrows with an ‘X’ indicate pseudogenes. Arrows with the same color adjacent to Tri genes are homologous genes based on between-species comparisons. Comparisons are derived from figures presented in Lee et al. [35] and Proctor et al. [33]. identification of ancestral duplications and the ability to pathogenicity. An important finding is that the most distinguish orthologous sequences from paralogous highly diverse genes found among strains of the same sequences. genus or even the same species, appear to be enriched for those involved in niche adaptation, including the coloni- Using this approach, a few fungal gene clusters respon- zation of living plant tissue. These genes or clusters often sible for the synthesis of plant-associated, bioactive com- are not randomly dispersed in the genome, but rather pounds have been shown to have likely evolved by way of tend to concentrate in genomic islands, telomere-prox- horizontal transfer. The sirodesmin/gliotoxin gene clus- imal regions or even entire lineage-specific chromosomes. ters that encode cyclic peptide derived toxins in several This organization and their association with mobile fungal species [40] and the ACE1 and ACE1-like gene genetic elements make them functionally analogous to clusters that encode enzymes apparently capable of genomic islands and plasmids in bacteria. synthesizing small molecules that can act as plant effectors [39,41] each show evidence for horiziontal transfer. What evolutionary processes organize genes involved in Additionally, as mentioned above, toxin gene clusters in pathogenicity and habitat adaptation into clusters and A. alternata may have been horizontally transferred by subgenomic regions? Answering this question will require virtue of their position on mobile supernumerary chromo- explanations for both mechanism and function. Do rates somes [21]. of mutation, recombination and duplication differ between conserved genomic regions and those associated Conclusions with plant interactions? Does chromatin modification play Comparative genomics greatly enhances the rate of a role in differences in the regulation of gene expression discovery of genes that form the basis of fungal between these regions? Is an increased level of organiz-

Current Opinion in Plant Biology 2010, 13:420–426 www.sciencedirect.com Genomic organization of plant pathogenicity in Fusarium species Rep and Kistler 425

ation, from pathogenicity gene cluster to pathogenicity Sub-telomere directed gene expression during initiation of invasive aspergillosis. PLoS Pathog 2008:4. chromosome, advantageous for the fungus? Are patho- 7. Gu¨ ldener U, Seong K, Boddu J, Cho S, Trail F, Xu JR, Adam G, genicity-related subgenomic regions more prone to repo- Mewes HW, Muehlbauer GJ, Kistler HC: Development of a sitioning within a genome or to mobility between Fusarium graminearum Affymetrix GeneChip for profiling fungal gene expression in vitro and in planta. Fungal Genet Biol genomes? Exactly what role do transposons play in gene 2006, 43:316-325. expression, mutation, recombination and mobility? The 8. de Wit P, Mehrabi R, van den Burg HA, Stergiopoulos I: Fungal answers to these questions, raised by comparative geno- effector proteins: past, present and future. Mol Plant Pathol mic analysis, will bring us closer to a unified understand- 2009, 10:735-747. ing of how fungal genomes adapt to a pathogenic 9. Stukenbrock EH, McDonald BA: Population genetics of fungal existence. and oomycete effectors Involved in gene-for-gene interactions. Mol Plant Microbe Interact 2009, 22:371-380. Acknowledgements 10. Gout L, Fudal I, Kuhn ML, Blaise F, Eckert M, Cattolico L, Balesdent MH, Rouxel T: Lost in the middle of nowhere: the We thank our colleague Li-Jun Ma for her enthusiasm and tireless efforts for AvrLm1 avirulence gene of the Dothideomycete a greater understanding of fungal biology through genomics. We also thank Leptosphaeria maculans. Mol Microbiol 2006, 60:67-80. Dr. Ma and Leslie Gaffney for the preparation of Figure 1.MR acknowledges the support of the Royal Netherlands Academy of Arts and 11. Parlange F, Daverdin G, Fudal I, Kuhn ML, Balesdent MH, Blaise F, Sciences, the Netherlands Organisation for Scientific Research and the Grezes-Besset B, Rouxel T: Leptosphaeria maculans avirulence Centre for Biosystems Genomics. HCK acknowledges the support of the gene AvrLm4-7 confers a dual recognition specificity by the United States Department of Agriculture, National Institute of Food and Rlm4 and Rlm7 resistance genes of oilseed rape, and Agriculture, Agriculture and Food Research Initiative and the National circumvents Rlm4-mediated recognition through a single Science Foundation, Environmental Genomics program. amino acid change. Mol Microbiol 2009, 71:851-863. The most recent paper of a series by this group demonstrating the striking genomic context of avirulence genes in L. maculans. Unique of this case is References and recommended reading the dual specificity conferred by the avirulence gene and the dependence Papers of particular interest, published within the period of review, of recognition by one of the resistance genes on a single amino acid have been highlighted as: polymorphism. 12. Fudal I, Collemare J, Bohnert HU, Melayah D, Lebrun MH: of special interest Expression of Magnaporthe grisea avirulence gene ACE1 is of outstanding interest connected to the initiation of appressorium-mediated penetration. Eukaryot Cell 2007, 6:546-554. ¨ 1. Cuomo CA, Guldener U, Xu JR, Trail F, Turgeon BG, Di Pietro A, 13. Ka¨ mper J, Kahmann R, Bo¨ lker M, Ma LJ, Brefort T, Saville BJ, Walton JD, Ma LJ, Baker SE, Rep M et al.: The Fusarium Banuett F, Kronstad JW, Gold SE, Muller O et al.: Insights from graminearum genome reveals a link between localized the genome of the biotrophic fungal plant pathogen Ustilago polymorphism and pathogen specialization . Science 2007, maydis. Nature 2006, 444:97-101. 317:1400-1402. Linking a fungal genome sequence, genome-wide gene expression data 14. Orbach MJ, Farrall L, Sweigard JA, Chumley FG, Valent B: A during plant infection, SNP mapping and a DNA sequence-anchored telomeric avirulence gene determines efficacy for the rice genetic map, the authors identify high diversity regions of the fungal blast resistance gene Pi-ta. Plant Cell 2000, 12:2019-2032. genome associated with plant infection. These regions are marked by elevated SNP density and recombination frequency and are enriched in 15. Khang CH, Park SY, Lee YH, Valent B, Kang S: Genome genes for predicted secreted proteins and specifically expressed during organization and evolution of the AVR-Pita avirulence gene infection. family in the Magnaporthe grisea species complex. Mol Plant Microbe Interact 2008, 21:658-670. 2. Coleman JJ, Rounsley SD, Rodriguez-Carres M, Kuo A, Wasmann CC, Grimwood J, Schmutz J, Taga M, White GJ, 16. Yoshida K, Saitoh H, Fujisawa S, Kanzaki H, Matsumura H, Tosa Y, Zhou SG et al.: The genome of Nectria haematococca: Chuma I, Takano Y, Win J, Kamoun S et al.: Association genetics contribution of supernumerary chromosomes to gene reveals three novel avirulence genes from the rice blast fungal expansion. PLoS Genet 2009:5. pathogen Magnaporthe oryzae. Plant Cell 2009, 21:1573-1591. Three supernumerary chromosomes with distinct gene content are identi- An impressive combined application of association genetics and genome fied from the whole genome sequence of N. haematococca (F. solani) sequence analysis to identify candidate effector and novel avirulence including a pathogenicity chromosome associated with host specialization. genes in a plant pathogenic fungus. Revealed high frequency of pre- sence/absence of candidate effector genes, association of genes for 3. Ma LJ, van der Does HC, Borkovich KA, Coleman JJ, secreted proteins and polymorphic genes with transposons, and con- Daboussi MJ, Di Pietro A, Dufresne M, Freitag M, Grabherr M, firmed the tendency of avirulence genes to reside in unstable genomic Henrissat B et al.: Comparative genomics reveals mobile regions. pathogenicity chromosomes in Fusarium. Nature 2010, 464:367-373. 17. He C, Rusu AG, Poplawski AM, Irwin JAG, Manners JM: Transfer of a supernumerary chromosome between vegetatively 4. Gale LR, Bryant JN, Calvo S, Giese H, Katan T, O’Donnell K, incompatible biotypes of the fungus Colletotrichum Suga H, Taga M, Usgaard TR, Ward TJ et al.: Chromosome gloeosporioides. Genetics 1998, 150:1459-1466. complement of the fungal plant pathogen Fusarium The first paper to experimentally demonstrate horizontal transfer of an graminearum based on genetic and physical mapping and entire fungal chromosome. cytological observations. Genetics 2005, 171:985-1001. 18. Lievens B, Houterman PM, Rep M: Effector gene screening 5. Fedorova ND, Khaldi N, Joardar VS, Maiti R, Amedeo P, allows unambiguous identification of Fusarium oxysporum Anderson MJ, Crabtree J, Silva JC, Badger JH, Albarraq A et al.: f.sp. lycopersici races and discriminationfrom other formae Genomic islands in the pathogenic filamentous fungus speciales. FEMS Microbiol Lett 2009, 300:201-215. Aspergillus fumigatus. PLoS Genet 2008:4. First comparative genomics study of a group of very closely related but 19. Temporini ED, VanEtten HD: An analysis of the phylogenetic phenotypically dissimilar filamentous fungi, including an important human distribution of the pea pathogenicity genes of Nectria pathogen. Revealed large species-specific gene sets, implicated in haematococca MPVI supports the hypothesis of their origin by habitat adaptation and concentrated in genomic islands enriched for horizontal transfer and uncovers a potentially new pathogen repetitive elements and pseudogenes. Genomic islands are proposed to of garden pea: Neocosmospora boniensis. Curr Genet 2004, serve simultaneously as gene dumps and gene factories. 46:29-36. 6. McDonagh A, Fedorova ND, Crabtree J, Yu Y, Kim S, Chen D, 20. Rodriguez-Carres A, White G, Tsuchiya D, Taga M, VanEtten HD: Loss O, Cairns T, Goldman G, Armstrong-James D et al.: The supernumerary chromosome of Nectria haematococca www.sciencedirect.com Current Opinion in Plant Biology 2010, 13:420–426 426 Biotic interactions

that carries pea-pathogenicity-related genes also carries a 31. Carbone I, Jakobek JL, Ramirez-Prado JH, Horn BW: trait for pea rhizosphere competitiveness. Appl Environ Recombination, balancing selection and adaptive evolution in Microbiol 2008, 74:3849-3856. the aflatoxin gene cluster of Aspergillus parasiticus. Mol Ecol 2007, 16:4401-4417. 21. Akagi Y, Akamatsu H, Otani H, Kodama M: Horizontal chromosome transfer, a mechanism for the evolution and 32. Maier FJ, Miedaner T, Hadeler B, Felk A, Salomon S, Lemmens M, differentiation of a plant-pathogenic fungus. Eukaryot Cell Kassner H, Schafer W: Involvement of trichothecenes in 2009, 8:1732-1738. fusarioses of wheat, barley and maize evaluated by gene Detailed analysis of a conditionally dispensable chromosome (CDC) disruption of the trichodiene synthase (Tri5) gene in three field associated with pathogenicity towards tomato of Alternaria alternata. isolates of different chemotype and virulence. Mol Plant Pathol High sequence similarity of genes on this chromosome between different 2006, 7:449-461. strains suggests horizontal transfer of pathogenicity chromosomes between strains. This possibility was supported by the ability to transmit 33. Proctor RH, McCormick SP, Alexander NJ, Desjardins AE: the CDC chromosome to another strain by protoplast fusion, resulting in a Evidence that a secondary metabolic biosynthetic gene strain with an expanded host range. cluster has grown by gene relocation during evolution of the filamentous fungus Fusarium. Mol Microbiol 2009, 22. Hatta R, Ito K, Hosaki Y, Tanaka T, Tanaka A, Yamamoto M, 74:1128-1142. Akimitsu K, Tsuge T: A conditionally dispensable chromosome controls host-specific pathogenicity in the fungal plant 34. Kimura M, Tokai T, O’Donnell K, Ward TJ, Fujimura M, pathogen Alternaria alternata. Genetics 2002, 161:59-70. Hamamoto H, Shibata T, Yamaguchi I: The trichothecene biosynthesis gene cluster of Fusarium graminearum F15 23. van der Does HC, Rep M: Virulence genes and the evolution of contains a limited number of essential pathway genes and host specificity in plant-pathogenic fungi. Mol Plant Microbe expressed non-essential genes. FEBS Lett 2003, 539:105-110. Interact 2007, 20:1175-1182. 35. Lee T, Han YK, Kim KH, Yun SH, Lee YW: Tri13 and Tri7 24. Gout L, Kuhn ML, Vincenot L, Bernard-Samain S, Cattolico L, determine deoxynivalenol-and nivalenol-producing Barbetti M, Moreno-Rico O, Balesdent MH, Rouxel T: Genome chemotypes of Gibberella zeae. Appl Environ Microbiol 2002, structure impacts molecular evolution at the AvrLm1 68:2148-2154. avirulence locus of the plant pathogen Leptosphaeria maculans. Environ Microbiol 2007, 9:2978-2992. 36. McCormick SP, Harris LJ, Alexander NJ, Ouellet T, Saparno A, 25. Wostemeyer J, Kreibich A: Repetitive DNA elements in fungi Allard S, Desjardins AE: Tri1 in Fusarium graminearum encodes (Mycota): impact on genomic architecture and evolution. a P450 oxygenase. Appl Environ Microbiol 2004, 70:2044-2051. Curr Genet 2002, 41:189-198. 37. McCormick SP, Alexander NJ, Proctor RH: Heterologous 26. Slotkin RK, Martienssen R: Transposable elements and the expression of two trichothecene P450 genes in Fusarium epigenetic regulation of the genome. Nat Rev Genet 2007, verticillioides. Can J Microbiol 2006, 52:220-226. 8:272-285. 38. Rosewich UL, Kistler HC: Role of horizontal gene transfer in the 27. Friesen TL, Stukenbrock EH, Liu ZH, Meinhardt S, Ling H, Faris JD, evolution of fungi. Annu Rev Phytopathol 2000, 38:325-363. Rasmussen JB, Solomon PS, McDonald BA, Oliver RP: Emergence of a new disease as a result of interspecific 39. Khaldi N, Collemare J, Lebrun MH, Wolfe KH: Evidence for virulence gene transfer. Nat Genet 2006, 38:953-956. horizontal transfer of a secondary metabolite gene cluster between fungi. Genome Biol 2008:9. 28. Keller NP, Hohn TM: Metabolic pathway gene clusters in filamentous fungi. Fungal Genet Biol 1997, 21:17-29. 40. Patron NJ, Waller RF, Cozijnsen AJ, Straney DC, Gardiner DM, Nierman WC, Howlett BJ: Origin and distribution of 29. Ward TJ, Bielawski JP, Kistler HC, Sullivan E, O’Donnell K: epipolythiodioxopiperazine (ETP) gene clusters in filamentous Ancestral polymorphism and adaptive evolution in the ascomycetes. BMC Evol Biol 2007:7. trichothecene mycotoxin gene cluster of phytopathogenic Fusarium. Proc Natl Acad Sci USA 2002, 99:9278-9283. 41. Collemare J, Billard A, Boehnert HU, Lebrun MH: Biosynthesis of secondary metabolites in the rice blast fungus Magnaporthe 30. Moore GG, Singh R, Horn BW, Carbone I: Recombination and grisea: the role of hybrid PKS-NRPS in pathogenicity. Mycol lineage-specific gene loss in the aflatoxin gene cluster of Res 2008, 112:207-215. Aspergillus flavus. Mol Ecol 2009, 18:4870-4887.

Current Opinion in Plant Biology 2010, 13:420–426 www.sciencedirect.com